AWS Polly (Text 2 Speech) - Easy install and usage + Multiple languages!

Hey everyone,

I wanted to let you know about a new project that I created for FreePBX called FreePBX-AWS-Polly. I got tired of the lack of good documentation on this topic and the scripts that were floating around in the forum were a bit sketchy as they used the AWS SDK 2, which is now deprecated, and they used Node 8.

So, I created a project that’s easy to install, bundled with Node 16, and uses AWS SDK 3. With FreePBX-AWS-Polly, you can easily use Amazon Polly to generate speech output in your FreePBX system in a variety of languages and voices.

The project includes a detailed installation guide and documentation on how to use the software and customize its settings. I encourage anyone who is interested in contributing to the project or has any questions to join us on the GitHub repository or the FreePBX community forum.

So, what do you think? Give it a try and let me know your thoughts. I’m excited to hear what you create with FreePBX-AWS-Polly!

github.com/scolastico/freepbx-aws-polly

or go directly to the step by step installation:
github.com/scolastico/freepbx-aws-polly/wiki/[ENGLISH]-Step-by-Step-guide-on-how-to-install.

2 Likes

Thanks for doing this. Built in TTS has been lacking for FreePBX and while we have hobbled solutions together, this is so clean.

Have you considered making a module for speech to text (speech to text) using Amazon? That would be a great complementary service to what you’ve already created and would really open up new interactive opportunities.

Yes im allready thinking about to integrate this inside a module so installation is a pice of cake also im thinking about integrating then a new panel in the system recordings tab so we can easly create tts system recordings.

Edit: Oh i misread, you meant for s2t. Sorry currently i dont have plans for that but eventually in the future.

1 Like

Seconded! If you are able to get to it s2t at some point in the future that would be an amazing feature!

I was reading though the github and one question I had was can you insert Asterisk variables into the TTS? “Hello, ${first_name}, thank you for calling…”

2 Likes

Im currently not sure if i’m able to introduce that feature but i like the idea. I will look into that the next days.

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.